Creative teams made up of both incumbents who knew each other and newcomers who didn’t, however, were more likely to produce hit shows. Uzzi says that’s because teams with too many overlaps in their social networks are less creative — the team members all know the same stuff. Teams that aren’t networked at all, however, aren’t good at sharing what they do know. The most successful teams were those in which everyone knew one or two others but not everyone — and not no one.

Moreover, innovation and creativity are more likely to blossom when people are exposed to various ideas and knowledge that are outside of their narrow expertise.  This is more likely to happen in a network with a higher percentage of weaker ties, versus overlapping redundancy:

“Creativity depends upon a person’s ability to take ideas and information that may be well understood in one area and bring it into a new area where it’s suddenly received as invention,” Uzzi says. Therein lies the productivity potential of social networks — they take whatever useful material is circulating and put it together in new ways. “Much of creativity is just new combinations put together from different pieces of information and material,” Uzzi says. “That’s how networks can really amplify creativity.”

For that reason, Uzzi suggests that organizations subvert the “proximity principle,” or people’s tendency to create networks from those around them. “The problem with the proximity principle is it tends to create homogeneous networks that lack diversity,” Uzzi says. “To undo the proximity principle . . . locate people from different specialties in the same area rather than keeping all specialists located near each other.”

Bottom line: businesses should think about how to optimally organize their workers for innovation.  Critical to this is encouraging what I would term “social bumping“–the process of being exposed to new and diverse ideas on a frequent basis.  If work environments are organized in such a way where workers only interact with people they share strong ties with they are less likely to be exposed to new and diverse ideas (since those we share strong ties with typically have access to the same knowledge and interests as we do).  The key is to leverage–and increase access to–the weak ties of workers, since these ties are the pipelines of new and diverse knowledge.  As mentioned in an earlier post, this can include helping people build diverse social networks, promoting the co-mingling of various subject-matter experts (e.g. internal rotations, inter-firm collaborations), leveraging social network technologies that encourage broad reading and sharing of knowledge (particularly the knowledge that flows from weak ties), and encouraging people to share their ideas and perspectives when they might otherwise feel as though it wasn’t their place to speak up.

Christakis and Fowler claim that social networks can be used as sensors for various types of contagions (whether biological, psychological, informational, etc).  In an inventive twist, they leverage what is known as the Friendship Paradox–the idea that, for almost everyone, a person’s friends tend to have more friends than they do.   Contagions tend to appear sooner in those individuals that are closer to the center of a social network.  The logic goes that if you ask a group of people to name one of their friends, those friends will be closer to the center of the network than the people you asked.  Rather than map and monitor an entire social network, simply monitoring these friends should allow researchers to detect the outbreak of, say, H1N1 much earlier.

They tested their theory using Harvard College undergrads, attempting to detect the outbreak of the flu.  (You can watch Christakis discuss the paper and research during a recent TED talk in the video embed below).  What did they find?

Based on clinical diagnoses, the progression of the epidemic in the friend group occurred 13.9 days (95% C.I. 9.9–16.6) in advance of the randomly chosen group (i.e., the population as a whole). The friend group also showed a significant lead time (p,0.05) on day 16 of the epidemic, a full 46
days before the peak in daily incidence in the population as a whole. This sensor method could provide significant additional time to react to epidemics in small or large populations under surveillance. The amount of lead time will depend on features of the outbreak and the network at hand. The method could in principle be generalized to other biological, psychological, informational, or behavioral contagions that spread in networks.

That is a pretty impressive result.  By simply tracking those individuals located closer to the center of the network, Christakis and Fowler were about to detect the progression of the flu a full 2 weeks before the general population.  They were also able to derive an early warning signal over a month before the peak of the outbreak in the general population.

If this result can be replicated and validated there are various ways it can be utilized.

Here are a few off the top of my head:

1. Product Launches: Particularly in the tech industry–where so often we now see product launches as proto-typing–we could use this method to very quickly gauge the awareness and adoption of a new product and predict the extent to which it will spread throughout the general population.  Companies would have better early-warning systems, which would allow for killing dud products or boosting marketing for those products that are poised to explode.  I would assume this would be particularly applicable to products that benefit/rely on network effects.
2. Political Indicators: One can think of political unrest as a contagion–discontent starting earlier with a core group within a social network and then, over time, spreading to those on the outskirts of the network.  Tracking the population as a whole may not give you an early warning of unrest, but rather a snapshot of a problem at a time when it is too late to do much about it.  Focusing on those closer to the core of a social network could provide enough lead time to diffuse tensions or intervene in other ways to avoid a full-scale upheaval.  Moreover, businesses and investors could also use the early warning as a signal to make adjustments in supply chain and their portfolios to take into account the potential unrest.  Finally, citizens within those countries could benefit by having more lead time to evacuate conflict zones, etc.
3. Economic Indicators: Investors, businesses, and politicians are always looking for better economic indicators–those signals that are leading indicators of larger economic trends.  I wonder if adjusting the sampling frames of various polls to incorporate the Friendship Paradox might give us an even earlier warning for mortgage defaults, consumer confidence and spending, manufacturing activity, etc.  Not as sure about this one, but certainly much of economic activity takes place in a networked structure.

Would love to hear other thoughts.

A recent paper by scholars working in Yahoo! Research looked at the extent to which friends on Facebook could accurately predict the degree to which their friends (placed on a continuum of strong to weak ties) agreed with them on various political issues.  The results:

We found that when friends disagree on a political issue, they are unaware of that disagreement about 60% of the time. Even close friends who discuss politics are typically unaware of their differences in opinions.

On the one hand, the result doesn’t surprise me all that much since political issues are arguably the most sensitive topic amongst friends and family and people tend to adjust their actions in order to fit into a socially-expected perception of who they are.  However, I would have guessed that people who interact on Facebook would be more likely than a control group to accurately predict the degree of agreement with friends.  For one, the platform is built to allow people to share their thoughts on any topic, so it provides a window into people’s feelings that face to face interaction may not.  Also, people tend to be less inhibited when there is physical distance between themselves and their interlocutors, so you would assume that topics that may be uncomfortable to discuss in person will be discussed on Facebook for all to see.  That makes the finding even more interesting for me.

For me, the most interesting finding is the disproportionate way that misery spreads in relation to happiness:

In the current study, Hill’s team compared patterns of relationships and emotions measured in the study to those generated by a model designed to track SARS, foot-and-mouth disease and other traditional contagions. They discounted spontaneous or immediately shared emotion — friends or relatives undergoing a common experience — and focused on emotional changes that followed changes in others.

In the spread of happiness, the researchers found clusters of “infected” and “uninfected” people, a pattern considered a “hallmark of the infectious process,” said Hill. “For happiness, clustering is what you expect from contagion rates. Whereas for sadness, the clusters were much larger than we’d expect. Something else is going on.”

Happiness proved less social than sadness. Each happy friend increased an individual’s chances of personal happiness by 11 percent, while just one sad friend was needed to double an individual’s chance of becoming unhappy. (my emphasis)

Something to think about.

Nicholas Christakis is a leading researcher focusing on the ways in which the social networks that we are embedded in (friends, family, work, etc) directly influence our behavior, health, and happiness.  Christakis recently presented at TED in February and the video was just made available this month:

Full disclosure: I have not read Christakis’ work–books, articles or otherwise (although his book Connected is on my list).  However, that won’t stop me from commenting on it.  Conceptually, I think this is fascinating research, particularly because modern technology allows for our social networks to be much larger and for us to have even greater exposure and interaction with first-degree members of our network, potentially amplifying the larger network’s influence on our behavior.  Granted, if you reduce the hypothesis to its simplest form–that people are influenced by those in their social circle–it seems somewhat obvious.  However, as with many things that we assume to be true the real revelation here is understanding the specific dynamics and mechanisms that make it so.  A few questions/comments come to mind:

• Self-selection and causality: Determining which direction the causality arrow points in this area strikes me as very difficult.  In general, Christakis seems to be arguing that social networks are causing behavior in individuals.  However, he does note that self-selection (or what he refers to as homophily) could also be at play–e.g. people tend to select themselves into particular groups because they feel comfortable surrounding themselves with like-minded.  I don’t see why there couldn’t be multiple causal stories here where the direction may be different at different times or for different issues.  Understanding which direction causality flows is critical, though, if one is thinking of implementing public policies to deal with, say, public health issues.  Are you targeting individuals or the social networks?
• Free will versus environmental determinism: The research also brings up an interesting philosophical question–what is the balance, then, between an individual’s ability to exercise free will and the deterministic influence of their environment (in this case, their social network)?  Christakis’ research points to a probabilistic relationship versus pure determinism (e.g. you are 47% more likely to be obese if members of your social network are, rather than guaranteed to be), but the question is still a tricky one.
• Business relevance: It certainly isn’t a revelation that, for example, consumers’ buying decisions can be influenced by their friends and social crowd.  Much ink has been spilled touting the need to create social proofs of one’s product or targeting influencers who can spread the desire for a product like a virus through their social network.  However, as mentioned earlier, Christakis’s research could open up new avenues if it is successful in identifying the specific mechanisms (and causal direction) through which social networks affect individual behavior and vice-versa.  Marketers would have  a field day.

When Robert Menschel, a senior director at Goldman Sachs Group Inc., was considering deals involving large consumer companies such as Procter & Gamble, he would pick the brain of fellow club member Ed Meyer, the former chief executive of Grey Advertising.

“We were all young kids starting out, and it is easy when you are so involved in building your career to lose touch with other people who are outside your field,” says Mr. Menschel, who has been at Goldman Sachs for 55 years. “It helped me to understand why other people do what they do—which is important in life and in business. You don’t learn anything from talking to sameness.”

The Wednesday 10 comprised, at various points, more than 20 men; the goal was a number small enough to maintain intimacy yet large enough to ensure that at least 10 members would show up for each of the monthly Wednesday-night meetings. No more than two representatives of any one industry were permitted. The idea was to combat insularity, to keep the men connected to people and events outside their own professions.

The engineering of the group is particularly interesting: keep the group small enough so members could develop some sense of intimacy, but large enough to ensure decent and consistent attendance; and ensure that no single industry was overrepresented.  By ensuring diverse membership the group could benefit from an exposure to ideas and viewpoints from outside their day-to-day professional circles.  Additionally, the members would invite guest speakers for each session, further benefiting from a diverse membership that could tap into various experts from all sorts of fields.

There are, of course, questions to what extent the group succeeded in maximizing diversity and guarding against sameness.  However, the fact that these individuals in late 1950 were cognizant of these issues and tried to guard against them by organizing the group in a specific way is quite impressive and instructional for folks living in the age of social media.

Now Helprin does make some valid points; however, I think on balance the conclusions he reaches about new technologies and techniques is overblown and based on his own biases and misunderstandings.  I’d like to speak to a few of them below:

Mark Helprin (MH): Before there was copyright, there was very little incentive for people to actually write things and assemble information. With the development of copyright, all that has increased.

I agree with Helprin here and there is quite a lot of excellent research (particularly by economic historians like Douglass North) to back up the notion that the institution of copyright and associated legal protections such as property rights, patents, etc, led to an explosion in creativity and economic development in the Western world.  So, point for Helprin.

Mary Kuntz (MK): [...]  There’s a lot of activity online right now that’s all about collaboration and crowd sourcing. And some of what gets produced that way people call art. I gather that you would take issue with that?

MH: I would. Crowd sourcing—to me, the words are a nightmare. The great achievement of Western civilization, anyway, has been to end the collective approach to things that marked the early history of man.

In other words, you were defined as part of a group. You were a serf. You were a peasant. You were a slave or whatever. And then in medieval times, you were a member of a guild. And the rights of the individual didn’t really count. What happened with the Greeks and then with Roman law and then over a long, long period culminating in modern times is that we have refined the rights of the individual. Now, obviously collaboration can be very powerful, and it’s important, and we do things in concert with one another. And we advance science that way and many things. But there’s really nothing that can substitute for one mind and one voice.

In a single thought, Helprin does two things: 1) illustrates that he does not truly understand crowdsourcing, but sees it as a threat anyway (and he is surely not alone here); and 2) his entire premise (“nothing can substitute for one mind and one voice) is based on a myth (that of the lone or heroic innovator/creator), one that is contradicted by the two sentences immediately preceding it.

First of all, Helprin’s conceptual connection between crowdsourcing and collectivism is utterly misplaced and flatly wrong. Just because a process or endeavor involves input and cooperation amongst a large number of people it does not mean that one is backsliding against the liberalization of society. I guess Helprin thinks we should disband all those pesky corporations that publish and ship his books, as well as those retail chains that put them in the hands of readers. The brilliance that is Helprin would be the proverbial tree falling in the woods with no one to hear if not for the collective efforts (ideas, labor, etc) of numerous individuals. All crowdsourcing does is provide an alternative (and, in some cases, a better) method for organizing individuals to tackle a collective task or to gather input and insights necessary for the accomplishment of some goal. There is no requirement that by taking part in crowdsourcing one must give themselves over to some collective identity. In fact, many people who participate in crowdsourcing do so during their spare time, and as an alternative to their 9-5 occupation (as Jeff Howe has noted, the process taps into people’s spare cycles). One might argue that crowdsourcing actually provides people with greater outlets for creativity, since those that are electrical engineers by day get a chance to solve complex design problems that have baffled bench chemists at Fortune 500 companies.

Second, Helprin falls into a familiar trap which is basing his argument on the notion that great creativity and discoveries is the result of single, heroic individual. This is hardly the case and there are piles and piles of research illustrating how collaboration is key. Helprin himself notes that advances in, for example, science are dependent on intra- and inter-temporal collaboration. We are better able to tackle complex problems when we have the benefit of diverse input. Additionally, our ability to create and come up with advancements is always dependent on previous work (“standing on the shoulders of giants“) and leveraging or combining creations by others. Scott Berkun, in his wonderful Myths of Innovation, sums it up well:

It wasn’t until the 1500s and the rise of the Renaissance that Western cultures grew comfortable acknowledging people’s creative abilities and individual achievements.

[...] Today, years away form the Renaissance, we’re still attached to the myth of the lone inventors. We do recognize collaboration and partnerships, but we often fall back on tales of lone innovators as heroic figures for reasons of convenience. We insist on isolating credit and dismissing the importance of others. Patent law, by design, credits one or a handful of individuals, assuming not only that ideas are unique and seperable, which is dubious, but that individual names can be given legal ownership of ideas.

Helprin didn’t invent the English language, the novel, the short story, various themes used in his work, and I guarantee that he has on more than one occasion drawn inspiration directly or indirectly from the minds and works of others. To paraphrase Keynes, in the long run everything is derivative.

What cinches Helprin’s lack of perspective for me is this exchange:

MK: Well, let me ask you something. This is just a hypothetical. What if you were working on a new novel and you got stuck on the ending, you just couldn’t find a way to wrap it up and make the plot work. And let’s say you put the problem to a wide following of smart, engaged, intelligent Mark Helprin fans. And one of them or several of them working together came up with the perfect solution. And you used their idea, which, in fact, was better than anything you would come up with. Would the novel that resulted be any less valid?

MH: I don’t know what you mean exactly by valid. But what I can say is I would never do that. If you look at the history of literature, it’s never been done that way. In school, they have what they call “brainstorming,” which I think is a comic-book word. They have what they call “writing webs.” They sit and they criticize each other’s writing, sort of like people in the Soviet on a factory floor.

And this is not the way that it should be. And I guarantee you that the product of this will be far, far less valuable than the product of somebody straining to do his best and taking responsibility for it also.

It’s never been done, so why bother; brainstorming is akin to forced collectivization under the Soviets; and he knows for sure that nothing ‘collectively’ produced could be as good as something produced by an individual.

There’s lots more there so I encourage you to read or listen to the entire interview. As I said, there are moments where Helprin makes solid points (points I agree with), but to say I disagree with his view on crowdsourcing, social technology, and innovation would be a massive understatement.

Goffman's classic text

[Goffman's] fundamental contribution is how he slants his premise that we present a front in all our behavior before others. You have to understand that this posturing is real and pervasive, but rarely a consequence of out-and-out deception, or because we have succeeded in deceiving outselves. Usually we simply associate certain behaviors as appropriate in certain circumstances; some stylization is inherent in our interactions.

For instance, just as a certain attention to style–or a stubborn flouting of its demands–enters into the clothes we choose to wear in public, there is inherent artificiality in our choice of screen name on a social network (unless an account related to our real name happens to be available). And whatever we choose certainly expresses something we want to reveal about our nature. This doesn’t mean we are deceiving ourselves or others–we are being ourselves, but in a stylized manner.

Goffman’s approach certainly applies online, because our postings–even our instant messages–are more deliberate acts than our informal behaviors in real life. Although some participants play at being flippant and spontaneous on Facebook walls and microblogs, they must have greater consciousness of their effects on the viewer than most dinner table guests or concert attendees. Our online personas, therefore, conform even more closely to Goffman’s idea of everyday life than our everyday life does.

Goffman is by far one of my favorite social scientists and a major influence on my graduate research and my worldview in general.  I recommend the entire piece.

It’s true – knowledge “reshuffling” is key to innovation (or, as you said “when it comes to thinking, we need more disruption, not less.”) This reshuffling, however, creates a lot of “noise” (how do you know what you are looking for before you find it?), and ultimately accounts for less-than-efficient organization, because it creates crazy transaction costs.

And reduction of transaction costs is what made firms show up in the first place (in opposition to markets), as Ronald Coase would have said.

So, the very condition that’s critical for generation of new knowledge and/or recombining old and new knowledge is actually detrimental for efficiency. The question then is, how to combine organizational efficiency with innovation?

I think Ana is right to invoke Coase’s theory and to be concerned about drastically increasing transaction costs and it got me thinking.

My view: while historically firms were able to lessen the search costs associated with valuable information, this no longer applies.  Firms are no longer the most efficient and effective means for collecting and sourcing innovative ideas and insights.  Therefore, as Coase’s theory would predict, firms may need to go beyond their four walls an turn to the market, effectively outsourcing (or co-sourcing) a large portion of idea generation.  Additionally, firms can further reduce information and search costs by leveraging technologies that tag, catalog, and organize knowledge–both within and outside firms.

The basic logic behind traditional, integrated firms is that it was cheaper to bring the various elements of production under one roof.  The idea was that it made sense for firms to enter into long-term contracts with employees because it would eliminate the potentially massive transaction costs associated with, for example, search and information costs one finds in the market.  In theory, when transaction costs are less within a firm than within the broader market, firms will take those items (e.g. labor, technology, distribution, etc.) in-house.  When costs are less in the market, firms will choose to outsource those items.

What many firms are realizing now is that, in many cases, they don’t have the knowledge internally to problem solve and innovate effectively.  The rapid and profuse rise of Open Innovation initiatives by tradional R&D stalwarts provides evidence for this shift.  And even if they do have that knowledge internally, firms are quite inefficient when it comes to quickly locating and leveraging that knowledge.  Firms are typically not structured to efficiently leverage the vast knowledge base they have accumulated internally in the form of their employees, especially as they grow in size and stove-pipe various functions and departments.  It can take weeks, if not months, for firms to locate individuals that have specific knowledge sets that can contribute to a particular problem or project (I’ve seen it firsthand, working with various clients on this very issue).

Given that firms are no longer as efficient at collecting, organizing, and tapping into knowledge under their own roof there are three likely (and logical) outcomes:

1. Firms will continue to outsource this function to minds outside of their four walls.  This could take the form of Open Innovation, crowdsourcing, or partnerships with more specialized firms that have developed platforms for locating and taxonomizing expertise and knowledge holders;
2. Firms will invest more heavily in software and technology that will allow them to catalog the knowledge under their roof, making search and location of that knowledge less costly.  Internal social networks are one way, talent software is another.
3. Firms that fail to adapt to this new reality will fail to innovate as effectively as other firms that due adapt, becoming obsolete.

Ana also brought up a simpler question–how do you know what you are looking for before you find it?  To me, social media makes it quite easy to bump into new perspectives in a quasi-directional or directionaless way.  You can use it to casually challenge your own views and find new and interesting ideas, or you can use it to conduct a focused search for specific experts or knowledge communities.

Ana recently wrote an interesting piece about the impact of organizational structure and systems on performance.  In it, she cited a post I wrote on the role of polymaths in innovation.  The gist of my post was that there is evidence that individuals with a broad, diverse knowledge base contribute more to innovation than highly specialized “experts”.  Ana claimed that what is likely more important than polymaths is the connections between them.  On this point I would also agree.

Weak vs. Strong Ties in a Social Network

The discussion has now turned to how you structure an organization to maximize the likelihood of individuals bumping into others with different sets of knowledge. And away we go…

I wrote the other day that, for me, the big payoff of social media is that it both exposes people to perspectives and knowledge-sets that they normally would not come into contact with. This accomplishes two things: 1) it can help destabilize stale ways of thinking and 2) prevent such stasis from setting in going forward. Social media can also help bring individuals together to collectively problem solve and create. Social media can be as active or passive as one wants it to be. It can be engaged so that people “bump” into new and interesting people (case in point, Ana and I just bumped into each other via our blogs–we were not formally introduced or looking for each other or people with our respective skills, etc.), or it can be used to actively organize a diverse group of minds with the goal of solving a complex problem.

So I would say that to encourage innovative thinking we want to use social technologies to structure collaborations between diverse individuals both within and outside of organizations. For each problem, you would want to bring together differing perspectives from within the organization. There are now platforms that facilitate the crowdsourcing of internal expertise (e.g. Imaginatik), not necessarily crowds outside the organization. This would have the added advantage of preventing the kind of social network ossification (where we end up going back to the same people over and over again for insights) that Ana worries about. Additionally, we can and should add an external module to this as well. Having ready access to external experts further increases the possible diversity of opinion to draw from. This can be done by tapping into an expert network (such as Gerson Lehrman Group, the firm I work for) as well as leveraging external crowdsourcing platforms (e.g. Chaordix).

Another idea is to make social media and social technologies mandatory for all workers. You can set up internal networks and have those networks integrate with external networks to maximize the likelihood of “intellectual bumping”.

So there are some initial thoughts. What do you think?